Received 24 January 2002 Accepted 19 February 2002 Published online 25 April 2002
Early maternal effects mediated by immunity depend on sexual ornamentation of the male partner Nicola Saino1* , Raffaella Paola Ferrari1, Roberta Martinelli1, Maria Romano1, Diego Rubolini1 and Anders Pape Mù ller2 1
Dipartimento di Biologia, Sez. Zoologia Scienze Naturali, UniversitaÁ degli Studi di Milano, via Celoria 26, I-20133 Milano, Italy 2 Laboratoire d’ Ecologie Evolutive Parasitaire, CNRS UMR 7103, Universite Pierre et Marie Curie, 7 quai St Bernard, Case 237, F-75252 Paris Cedex 05, France Vertebrates have an immature immune system soon after birth, and parasites can therefore be particularly virulent to young hosts. Transfer of immune factors via the egg can give rise to early maternal effects with important consequences for offspring ® tness, as maternally derived immunity confers anti-parasite protection. Mothers are expected to allocate immunity differentially to the eggs according to the reproductive value of their offspring as in¯ uenced by the quality of their father. In this study, we analysed transmission to the yolk of antibodies speci® c to an antigen (Newcastle disease virus vaccine, NDV) by vaccinated female barn swallows (Hirundo rustica) mated to males whose secondary sexual characteristics had been manipulated. Concentration of anti-NDV antibodies in the yolk positively covaried with that in maternal plasma. Anti-NDV antibodies were more concentrated in the ® rst but not the fourth eggs laid by females mated with tail-elongated males compared with those mated with tail-shortened and control males. This experiment shows that allocation of maternal immune factors to the eggs is affected by quality of the male, as signalled by its secondary sexual characteristic. Thus, early maternal effects are in¯ uenced by sexual attractiveness of male mates and are mediated by immunity. Keywords: egg lay order; Hirundo rustica; immunity; maternal effects; secondary sexual characteristics
1. INTRODUCTION Vertebrates have evolved complex immune functions to defend themselves against parasites, which can have a strong negative impact on the ® tness of their hosts (Tizard 1991; Wakelin 1996; Mù ller 1997; Pastoret et al. 1998). Acquired immunity provides hosts with highly speci® c and ef® cient mechanisms to fend off parasites. However, acquired immune defences are weak and ensue slowly soon after birth (Apanius 1997; Pastoret et al. 1998). In addition, synchronization of a parasite reproductive cycle with that of the host (Noble & Noble 1976; Cox 1982), and particular behavioural characteristics of the host, such as coloniality, may increase offspring exposure to parasites (Loye & Zuk 1991; Clayton & Moore 1997). Females of oviparous vertebrates deliver to their eggs components of both innate and acquired immunity. This physiological adaptation can function as an early maternal effect (Mousseau & Fox 1998). Indeed, immune factors that are contained in the egg are passed on to the embryo and, consequently, to the newborn, and may thus enhance anti-parasite defence in the crucial early life stages, compensating for poor offspring immunocompetence (Naqi et al. 1983; Tizard 1991; Graczyk et al. 1994; Smith et al. 1994; Lung et al. 1996; Ahmad & Siddique 1998; Gasperini et al. 2001; see also Heeb et al. 1998). Resource allocation to and care of offspring involve individuals in costs in terms of survival or future repro*
Author for correspondence (
[email protected]).
Proc. R. Soc. Lond. B (2002) 269, 1005± 1009 DOI 10.1098/rspb.2002.1992
duction (LindeÂn & Mù ller 1989; Clutton-Brock 1991; Stearns 1992). The theory of parental care therefore predicts that strategies should have evolved that allow differential investment in relation to the quality of the offspring, as in¯ uenced for example by the quality of their father. According to some theoretical models of sexual selection, male ornaments have evolved under the in¯ uence of female preferences for traits that reliably reveal the quality of potential mates (reviewed in Andersson (1994)). Females with a directional preference for extreme ornaments may thus accrue indirect ® tness bene® ts in terms of inheritance of paternal genes for high viability or sexual attractiveness of the offspring. Hence, evolutionary theories of sexual selection and parental care predict that mothers should invest more in the production of offspring sired by high-quality males with large secondary sexual characteristics (Gil et al. 1999; Cunningham & Russell 2000, 2001; Sheldon 2000; Colegrave 2001; Gil & Graves 2001). The aim of this study was to investigate experimentally the relationship between the expression of a secondary sexual characteristic under directional female preference (length of the outermost tail feathers) of male barn swallows (Hirundo rustica) and concentration of antibodies speci® c to a novel antigen (Newcastle disease virus vaccine, NDV) in the yolk of the eggs laid by their mates. Females were vaccinated before start of laying, while males were subjected to either tail-shortening, tail-elongation or two control treatments. In a previous study we found that antiNDV antibodies are transmitted via the egg in this species,
1005
Ó
2002 The Royal Society
1006 N. Saino and others Antibody transfer to the eggs and male ornamentation
as they were detected in the progeny of immunized females 5 days after hatching (N. Saino, unpublished results). However, concentration of antibodies in the hatchlings may be affected by protein metabolism as in¯ uenced by nutritional condition, current infection and sib± sib competition. Therefore, we measured the effect of male ornamentation on antibody transfer to the yolk in unincubated eggs. We predicted that females would transfer a larger amount of anti-NDV antibodies to the eggs when their offspring were sired by males displaying experimentally enlarged sexual ornaments. The barn swallow is a semicolonial, socially monogamous, biparental passerine (Mù ller 1994). Females prefer males with long ornamental outermost tail feathers both as social and extra-pair mates (e.g. Mù ller 1994; Saino et al. 1997a). Long-tailed males are more viable, are capable of raising stronger immune responses and have smaller parasite loads than short-tailed males (Mù ller 1994; Saino & Mù ller 1996; Saino et al. 1997b, 1999a). 2. METHODS We studied barn swallows in four farms east of Milano (northern Italy) during the breeding season of 2001. We captured adult swallows as soon as they started visiting the rural buildings where they breed. At ® rst capture, adults were sexed, and length of male ornamental tail feathers was measured (Mù ller 1994). Each individual was given a combination of coloured leg rings and dye markings on the breast feathers that allowed assignment to nests by direct observation. A sample of blood was taken in heparinized capillary tubes. Males were assigned sequentially to one of the following tail manipulation treatments: males of the ® rst group had their outermost tail feathers shortened by 20 mm; males of the second group were captured and subjected to morphological measurements; males of the third group had their outermost tail feathers cut and re-glued without altering their length; and the tail of males of the fourth group was elongated by 20 mm (see Mù ller (1994) and Saino et al. (1997a,b) for methodological details). Before starting the assignment, males were arranged in a random order so that treatment sequence did not re¯ ect capture sequence. Tail length manipulation occurred three weeks on average (and never less than 10 days) before laying of the ® rst egg by the mate of the focal males, with no signi® cant difference in breeding stage at tail manipulation (number of days between manipulation and ® rst egg laying) among the four groups of males (ANOVA: F3,58 = 0.62, p = 0.61). Females were injected subcutaneously with 20 m l of NDV vaccine (Nobivac Paramyxo, Intervet, Milano, Italy). Blood samples were also collected 14 and 28 days post-vaccination. Plasma was then stored at 2 20 °C for assay of anti-NDV vaccine antibodies. All females were immunized more than 9 days before laying of their ® rst egg. Mean stage in the breeding cycle at NDV inoculation (number of days between inoculation and ® rst egg laying) was similar in the four experimental groups (tail-shortened: 16.6 (1.41 s.e.m.); unmanipulated: 17.1 (1.46 s.e.m.); cut: 16.0 (1.06 s.e.m.); tail-elongated: 15.9 (1.45 s.e.m.); ANOVA: F3,58 = 0.44, p = 0.72). Nests were inspected every day and eggs were marked. We collected the ® rst and fourth eggs from 14 clutches of tail-shortened males, 17 clutches of unmanipulated males, 16 clutches of males with cut and re-glued tail feathers and 15 clutches of tailelongated males. However, from four of these clutches the ® rst Proc. R. Soc. Lond. B (2002)
but not the fourth egg was available for immunological analyses (see ® gure 1). To minimize the impact on reproductive output of breeding colonies, we collected all the eggs only in a subset of the clutches from which we collected the ® rst egg (sample size: ® ve clutches for tail-shortened, four for unmanipulated, ® ve for cut and ® ve for tail-elongated males). This allowed us to check for consistency of variation in antibody concentration in the ® rst and fourth eggs, and also in the second and third eggs. However, in the analyses we mainly focused on the ® rst and fourth eggs. Eggs were collected on the second day after laying of the last egg, when the developing embryo is still very small, in order to be sure that clutch was complete. Each yolk was mixed and stored at 2 20 °C.
(a) Competitive enzyme-linked immunosorbent assay Anti-NDV antibody concentration was measured by monoclonal antibody-blocking enzyme-linked immunosorbent assay (ELISA) (Svanovir NDV-Ab, SVANOVA Biotech, Uppsala, Sweden) (Czifra et al. 1996; Nordling et al. 1998). Optical density (OD) values of test plasma were compared with the OD value of the kit NDV-negative control, while using plasma samples at the time of vaccination as a reference to assess the effectiveness of NDV vaccine injection in elicting a response. OD of test yolks were compared with both the kit NDV-negative control and with a pool of yolks from uninjected females. ODs were used to calculate percentage inhibition (PI) values. PI values of yolks were signi® cantly greater than zero when referred to the kit NDV-negative control and also to the pool of yolks ( p , 0.001 in both cases), indicating that vaccination resulted in antibody transfer to the eggs (see also § 3). Large PI values indicate large NDV-speci® c blocking antibodies in the plasma and yolk samples. Clutches were assigned to ELISA plates randomly. Within-plate repeatability of OD values, as determined on different plasma and yolk samples assayed in duplicate, was 0.92 and 0.93 (ANOVA: p , 0.003 in both cases). In the analyses of antibody concentration in eggs of different laying order, signi® cance values were corrected by a sequential Bonferroni procedure for four simultaneous tests on eggs of laying order 1 to 4. Post hoc Bonferroni tests for differences among experimental groups are presented where needed. Observed power of non-signi® cant statistical tests is reported.
3. RESULTS Vaccination caused an increase in anti-NDV antibodies in females, as determined by comparing PI of antigen activity at the time of inoculation with that recorded two weeks later (increase in PI: 20.5 (2.62 s.e.m.); paired t-test; t5 9 = 7.81, p , 0.0001). Antibody concentration increased further from day 14 to day 28 of vaccination (increase in PI: 14.29 (2.39 s.e.m.); t4 9 = 5.98, p , 0.0001). Female antibody concentration two weeks after vaccination was not affected by tail manipulation of her mate (ANOVA: F3 ,58 = 1.83, p = 0.15; power = 0.452), and the same held true four weeks after vaccination (ANOVA: F3 ,4 6 = 0.35, p = 0.79; power = 0.112), although the power of the latter test was low and thus the nonsigni® cant result could be due to a type II statistical error. There was no signi® cant difference in antibody levels among the experimental groups of females (repeatedmeasures ANOVA: between-subjects effect, F3 ,4 6 = 1.42, p = 0.25; power = 0.352) nor differential variation between
anti-NDV antibody concentration (PI)
Antibody transfer to the eggs and male ornamentation
50
5
40
5
30 20
15
4 5 4
14
17 16
5 5
10 0
first
13 15 15 15
5
second
third
fourth
egg lay order Figure 1. Mean (1 s.e.m.) anti-NDV antibody concentration in the ® rst to fourth eggs laid by females mated to males belonging to the four tail-manipulation treatments (white bars, shortened tail; horizontal lined bars, unmanipulated tail; vertical lined bars, cut and re-glued tail; black bars, elongated tail). Sample sizes are shown above bars. To minimize the impact on reproduction we collected second and third eggs only from a subset of the nests where we collected the ® rst and fourth eggs. Fifth and sixth eggs were neglected because of small sample sizes.
day 14 and day 28 among male treatments (F3 ,46 = 1.84, p = 0.15; power = 0.447). Antibody concentration in maternal plasma on day 28 positively covaried with that on day 14 (ANCOVA: F1 ,45 = 31.06, p , 0.0001). A signi® cant effect of male tail manipulation on antibody concentration was observed in ® rst eggs in an ANCOVA (F3 ,5 7 = 5.59, p , 0.01 after Bonferroni correction; ® gure 1), with a signi® cant positive covariation between antibody concentration in the eggs and original tail length of males (F1 ,57 = 8.32, p = 0.006, coef® cient = 0.45 (0.16 s.e.m.)). Concentration of antibodies in the ® rst eggs of females mated to tail-elongated males was signi® cantly larger than in each of the other groups ( post hoc Bonferroni test: p , 0.012 for all comparisons). However, females mated to unmanipulated, cut or tail-shortened males had non-signi® cantly different antibody concentration in their ® rst egg ( post hoc Bonferroni tests, p . 0.99 in all cases). We could ® nd no signi® cant effect of tail manipulation on antibody concentration in fourth eggs (ANCOVA: F3 ,53 = 0.17, p = 0.92, n.s., power = 0.079, ® gure 1; effect of original tail length: F1 ,53 = 2.47, p = 0.12, power = 0.339). Variation of antibody concentration between the ® rst and the fourth eggs differed among tail manipulation treatments (repeated-measures ANCOVA: interaction between lay order and tail treatment: F3 ,5 4 = 4.55, p = 0.006). Antibody concentration in the yolk of ® rst eggs positively covaried with concentration in the plasma of the mother measured two weeks post-vaccination, (ANCOVA: F1 ,5 6 = 5.59, p = 0.021, coef® cient = 0.14 (0.06 s.e.m.)), whereas there was no signi® cant evidence for a covariation between antibody concentration in fourth eggs and anti-NDV antibodies in maternal plasma (ANCOVA: F1 ,52 = 0.07, p = 0.79; power = 0.058). Covariation between clutch size and anti-NDV antibody concentration in the yolk of ® rst and fourth eggs was Proc. R. Soc. Lond. B (2002)
N. Saino and others 1007
not statistically signi® cant (repeated-measures ANCOVA: t = 2 0.42, p = 0.68). Among second and third eggs, for which a smaller sample was available, anti-NDV antibody concentration in the eggs laid by females of tail-shortened males ranked lowest while those of tail-elongated males ranked highest. In analyses of covariance with original tail length as covariate, antibody concentration in second eggs did not vary with tail treatment (F3 ,1 4 = 0.70, p = 0.57, power = 0.796), whereas variation was signi® cant among third eggs (F3 ,1 4 = 4.79, p = 0.016, p , 0.05 after Bonferroni correction). Concentration of antibodies in the third eggs of females mated to tail-elongated males was signi® cantly larger than in the eggs of females mated to tail-shortened males ( post hoc Bonferroni test: p = 0.012), whereas the other pairwise comparisons among groups of third eggs were not signi® cant ( p . 0.36 in all cases). All analyses described above on egg yolks were repeated using antibody concentration estimates based on negative controls consisting of a pool of yolks from unvaccinated females, and they gave qualitatively similar results. 4. DISCUSSION In this study we have shown that concentration of maternally derived antibodies in the egg yolk of barn swallows is in¯ uenced by the expression of a secondary sexual characteristic of the male partner. First eggs of females mated with tail-elongated males contained more antibodies than those laid by females of two control groups and those mated with tail-shortened mates, and a similar pattern of variation was observed in third eggs. These data were unaffected by physiological processing of yolk material during embryo development and variation of breeding stage at vaccination. The steady increase in antibody concentration until 28 days after vaccination suggests that at the time of egg laying, on average 16 days after vaccination, humoral response was still developing. The positive covariation between antibody levels in the mothers recorded 14 and 28 days post-vaccination, respectively, indicates that relative antibody levels at different times were consistent among females. To our knowledge, this is the ® rst experimental study showing a causal link between the expression of a secondary sexual characteristic of males and immunological quality of the eggs laid by their mates. Differential concentration of immune factors in the ® rst eggs may enhance immune protection of offspring sired by highquality males. In the barn swallow, ® rst eggs are known to produce offspring that are larger and thus more viable than later eggs (Mù ller 1994; Saino et al. 2001a). Immune defence is known to affect survival in barn swallows (Saino et al. 1997b), suggesting that eggs and nestlings that received more antibodies were more viable. Alternatively, a larger concentration of antibodies in the eggs of females mated to tail-elongated males may parallel variation in the offspring sex ratio associated with male ornamentation (Ellegren et al. 1996), as the immune pro® le of the offspring may have to be tuned according to their sex. In fact, variation in egg components has been shown to be related to offspring sex (Anderson et al. 1997; Cordero et al. 2000, 2001; Cunningham & Russell 2001). This intrepretation, however, is unlikely because in
1008 N. Saino and others Antibody transfer to the eggs and male ornamentation
another experiment there was no effect of tail manipulation on progeny sex ratio (Saino et al. 1999b). We found no signi® cant evidence of variation in maternal plasma anti-NDV antibodies among male tailmanipulation treatments. Hence, females mated with the most ornamented males may have allocated their larger production of antibodies to their ® rst eggs without increasing their own levels of immune defence. The fact that in fourth eggs no evidence of signi® cant variation in antibody concentration existed may indicate that a physiological constraint exists on the amount of antibodies that females can transfer to consecutive eggs, making it impossible for mates of tail-elongated males to allocate additional antibodies to their last eggs. However, the power of this statistical test was low, indicating that results on fourth eggs may have been affected by a type II statistical error. Present results indicate that females did not just passively transfer antibodies to their eggs, as this hypothesis would predict similar responses independently of male ornamentation and lay order. In addition, females did not adopt an indiscriminate allocation strategy relative to mate attractiveness, as antibody concentration depended on lay order. This indicates that females follow a ® ne-tuned strategy of allocation depending on male ornamentation and lay order. Because ® fth and sixth eggs were rare, we can exclude the possibility that females simply allocated a ® xed amount of antibody to a clutch, because females mated to tail-elongated males allocated more antibody to their clutch than any other female (® gure 1). The results of this study have obvious implications for theories of the evolution of parental investment and sexual selection. Transfer of passive immunity can have important consequences for condition and survival of offspring later on (see § 1). However, females may experience a cost of reproduction consisting of biosynthesis of antibodies, as also shown by the increase in total antibody concentration of female barn swallows prior to egg laying (Saino et al. 2001b). Costs of immune system functioning (Klasing & Leshchinsky 1999) and parasitism may select for the evolution of allocation strategies that allow females to invest more in immunological protection of offspring with large reproductive value. If males with large sexual ornaments are sexually attractive and/or reliably signal their desirable genetic quality, females should value their offspring more when sired by such males. In addition, the positive covariation we observed between pre-manipulation length of male tail ornaments and antibody concentration in the ® rst eggs indicates that egg quality was in¯ uenced by original male phenotype, possibly because females responded to male traits (e.g. carotenoid-based coloration of forehead and throat feathers) that are correlated with length of ornamental tail feathers. We conclude that early maternal effects mediated by the immune system vary in relation to the quality of male mates, as re¯ ected by the expression of a secondary sexual characteristic, in accordance with evolutionary theory predicting larger investment in offspring with large reproductive value. We thank T. A. Mousseau and M. Petrie for constructive comments on a draft. Proc. R. Soc. Lond. B (2002)
REFERENCES Ahmad, M. & Siddique, M. 1998 Kinetics of antibodies in serum, egg yolk and day-old chicks against infectious bursal disease in chicken broiler breeders. Pakistan Vet. J. 18, 187± 191. Anderson, D. J., Reeve, J. & Bird, D. M. 1997 Sexually dimorphic eggs, nestling growth and sibling competition in American kestrels Falco sparverius. Funct. Ecol. 11, 331± 335. Andersson, M. 1994 Sexual selection. Princeton University Press. Apanius, V. 1997 The immune system. In Avian growth and development: evolution within the altricial± precocial spectrum, Oxford Ornithological Series, no. 8. (ed. J. M. Starch & R. E. Ricklefs), pp. 130± 145. Oxford University Press. Clayton, D. H. & Moore, J. (eds) 1997 Host± parasite evolution: general principles and avian models. Oxford University Press. Clutton-Brock, T. H. 1991 The evolution of parental care. Princeton University Press. Colegrave, N. 2001 Differential allocation and `good genes’ Ð comment from Colegrave. Trends Ecol. Evol. 16, 22± 23. Cordero, P. J., Grif® th, S. C., Aparicio, J. M. & Parkin, D. T. 2000 Sexual dimorphism in house sparrows’ eggs. Behav. Ecol. Sociobiol. 48, 353± 357. Cordero, P. J., Vinuela, J., Aparicio, J. M. & Veiga, J. P. 2001 Seasonal variation in sex ratio and sexual egg dimorphism favouring daughters in ® rst clutches of the spotless starling. J. Evol. Biol. 14, 829± 834. Cox, F. E. G. 1982 Modern parasitology. Oxford: Blackwell. Cunningham, E. J. A. & Russell, A. F. 2000 Egg investment is in¯ uenced by male attractiveness in the mallard. Nature 404, 74± 76. Cunningham, E. J. A. & Russell, A. F. 2001 Differential allocation and `good genes’ Ð comment from Cunningham & Russell. Trends Ecol. Evol. 16, 21. Czifra, G., Nilsson, M., Alexander, D. J., Manvell, R., KecskemeÂti, S. & EngstroÈm, B. E. 1996 Detection of PMV-1 speci® c antibodies with a monoclonal antibody blocking enzyme-linked immunosorbent assay. Avian Pathol. 25, 691± 703. Ellegren, H., Gustafsson, L. & Sheldon, B. C. 1996 Sex ratio adjustment in relation to parental attractiveness in a wild bird population. Proc. Natl Acad. Sci. USA 93, 11 723± 11 728. Gasperini, J., McCoy, K. D., Haussy, C., Tveraa, T. & Boulinier, T. 2001 Induced maternal response to the Lyme disease spirachaete Borrelia burgdorferi sensu lato in a colonial seabird, the kittiwake Rissa tridactyla. Proc. R. Soc. Lond. B 268, 647± 650. Gil, D. & Graves, J. 2001 Differential allocation and `good genes’ Ð comment from Gil & Graves. Trends Ecol. Evol. 16, 21± 22. Gil, D., Graves, J., Hazon, N. & Wells, A. 1999 Male attractiveness and differential testosterone investment in zebra ® nch eggs. Science 286, 126± 128. Graczyk, T. K., Cran® eld, M. R., Shaw, M. L. & Craig, L. E. 1994 Maternal antibodies against Plasmodium spp. in African black-footed penguin chicks. J. Wildl. Dis. 30, 365± 371. Heeb, P., Werner, I., KoÈlliker, M. & Richner, H. 1998 Bene® ts of induced host responses against an ectoparasite. Proc. R. Soc. Lond. B 265, 51± 56. Klasing, K. C. & Leshchinsky, T. V. 1999 Functions, costs, and bene® ts of the immune system during development. In Proc 22nd Int. Orn. Congr. (ed. N. Adams & R. Slotow), pp. 2817± 2835. Durban, South Africa: University of Natal. LindeÂn, M. & Mù ller, A. P. 1989 Cost of reproduction and covariation of life history traits in birds. Trends Ecol. Evol. 4, 367± 371.
Antibody transfer to the eggs and male ornamentation Loye, J. E. & Zuk, M. 1991 Bird± parasite interactions. Oxford University Press. Lung, N. P., Thompson, J. P., Collias, G. V., Olsen, J. H., Zdziarski, J. A. & Klein, P. A. 1996 Maternal transfer of IgG antibodies and development of IgG antibody responses by blue and gold macaw chicks (Ara ararauna). Am. J. Vet. Res. 57, 1157± 1161. Mù ller, A. P. 1994 Sexual selection and the barn swallow. Oxford University Press. Mù ller, A. P. 1997 Parasitism and the evolution of host life history. In Host± parasite evolution: general principles and avian models (ed. D. H. Clayton & J. Moore), pp. 105± 127. Oxford University Press. Mousseau, T. A. & Fox, C. W. (eds) 1998 Maternal effects. New York: Oxford University Press. Naqi, S. A., Marquez, B. & Sahin, N. 1983 Maternal antibody and its effect on infectious bursal disease virus. Avian Dis. 25, 831± 838. Noble, E. R. & Noble, G. A. 1976 Parasitology, 4th edn. Philadelphia, PA: Lea & Fabiger. Nordling, D., Andersson, M., Zohari, S. & Gustafsson, L. 1998 Reproductive effort reduces speci® c immune response and parasite resistance. Proc. R. Soc. Lond. B 265, 1291± 1298. Pastoret, P., Gabriel, P., Bazin, H. & Govaerts, A. 1998 Handbook of vertebrate immunology. San Diego, CA: Academic. Saino, N. & Mù ller, A. P. 1996 Sexual ornamentation and immunocompetence in the barn swallow. Behav. Ecol. 7, 227± 232. Saino, N., Primmer, C. R., Ellegren, H. & Mù ller, A. P. 1997a An experimental study of paternity and tail ornamentation in the barn swallow (Hirundo rustica). Evolution 51, 562± 570.
Proc. R. Soc. Lond. B (2002)
N. Saino and others 1009
Saino, N., Bolzern, A. M. & Mù ller, A. P. 1997b Immunocompetence, ornamentation and viability of male barn swallows (Hirundo rustica). Proc. Natl Acad. Sci. USA 97, 579± 585. Saino, N., Calza, S., Ninni, P. & Mù ller, A. P. 1999a Barn swallows trade survival against offspring condition and immunocompetence. J. Anim. Ecol. 68, 999± 1009. Saino, N., Ellegren, H. & Mù ller, A. P. 1999b No evidence for adjustment of sex allocation in relation to paternal ornamentation and paternity in barn swallows. Mol. Ecol. 8, 399± 406. Saino, N., Incagli, M., Martinelli, R., Ambrosini, R. & Mù ller, A. P. 2001a Immunity, growth and begging behaviour of nestling barn swallows Hirundo rustica in relation to hatching order. J. Avian Biol. 32, 263± 270. Saino, N., Martinelli, R. & Mù ller, A. P. 2001b Immunoglobulin plasma concentration in relation to egg laying and mate ornamentation of female barn swallows (Hirundo rustica). J. Evol. Biol. 14, 96± 104. Sheldon, B. C. 2000 Differential allocation: tests, mechanisms and implications. Trends Ecol. Evol. 15, 397± 402. Smith, N. C., Wallach, M., Miller, C. M. D., Morgenstern, R., Braun, R. & Eckert, J. 1994 Maternal transmission of immunity to Eimeria maxima: enzyme-linked immunosorbent assay analysis of protective antibodies induced by infection. Infect. Immun. 62, 1348± 1357. Stearns, S. C. 1992 The evolution of life histories. Oxford University Press. Tizard, I. 1991 Veterinary immunology. Philadelphia, PA: W.B. Saunders. Wakelin, D. 1996 Immunology to parasites. Cambridge University Press.
